FOXC1 promotes HCC proliferation and metastasis by Upregulating DNMT3B to induce DNA Hypermethylation of CTH promoter

Abstract

Background: Forkhead box C1 (FOXC1), as a member of the FOX family, is important for promote HCC invasion and metastasis. FOX family protein lays a pivotal role in metabolism. ROS is involved in tumor progression and is associated with the expression of lots of transcription factors. We next explored the mechanism underlying FOXC1 modulating the metabolism and ROS hemostasis in HCC. Methods: We used amino acids arrays to verify which metabolism is involved in FOXC1-induced HCC. The kits were used to detect the ROS levels in HCC cells with over-expression or down-expression of FOXC1. After identified the downstream target genes and candidate pathway which regulated by FOXC1 during HCC progression in vitro and in vivo, we used western blot, immunohistochemistry, bisulfite genomic sequencing, methylation-specific PCR, chromatin immunoprecipitation analysis and luciferase reporter assays to explore the relationship of FOXC1 and downstream genes. Moreover, the correlation between FOXC1 and target genes and the correlation between target genes and the recurrence and overall survival were analyzed in two independent human HCC cohorts. Results: Here, we reported that FOXC1 could inhibit the cysteine metabolism and increase reactive oxygen species (ROS) levels by regulating cysteine metabolism-related genes, cystathionine γ-lyase (CTH). Overexpression of CTH significantly suppressed FOXC1-induced HCC proliferation, invasion and metastasis, while the reduction in cell proliferation, invasion and metastasis caused by the inhibition of FOXC1 could be reversed by knockdown of CTH. Meanwhile, FOXC1 upregulated de novo DNA methylase 3B (DNMT3B) expression to induce DNA hypermethylation of CTH promoter, which resulted in low expression of CTH in HCC cells. Moreover, low levels of ROS induced by N-acetylcysteine (NAC) which is an antioxidant inhibited the cell proliferation, migration, and invasion abilities mediated by FOXC1 overexpression, whereas high levels of ROS induced by L-Buthionine-sulfoximine (BSO) rescued the suppression results mediated by FOXC1 knockdown. Our study demonstrated that the overexpression of FOXC1 that was induced by the ROS dependent on the extracellular regulated protein kinases 1 and 2 (ERK1/2)- phospho-ETS Transcription Factor 1 (p-ELK1) pathway. In human HCC tissues, FOXC1 expression was positively correlated with oxidative damage marker 8-hydroxy-2′-deoxyguanosine (8-OHdG), p-ELK1 and DNMT3B expression, but negatively correlated with CTH expression. HCC patients with positive co-expression of 8-OHdG/FOXC1 or p-ELK1/FOXC1 or FOXC1/DNMT3B had the worst prognosis, whereas HCC patients who had positive FOXC1 and negative CTH expression exhibited the worst prognosis. Conclusion: In a word, we clarify that the positive feedback loop of ROS-FOXC1-cysteine metabolism-ROS is important for promoting liver cancer proliferation and metastasis, and this pathway may provide a prospective clinical treatment approach for HCC.